I am working with my favorite collaborator, Brian Lamb, on a position paper for a conference at the Open University of Catalonia this fall around the topic: “Pushing the boundaries of Higher Education: challenging traditional models with innovative and creative practices.” This is my early contribution, and Brian will follow soon which will fine tune some of the critiques here as well highlight interventions from various folks.
I have been talking a bit about the EDUCAUSE Next Generation Digital Learning Environment (NGDLE) paper from 2015 in my talks recently as a one possible vision of how loosely coupled publishing platform connecting various tools could be one way to imagine the power of what Kin Lane defines as the Personal API, which frames the importance of getting individuals more control over who and what has access to their online data. The learning management system (LMS or VLE in the UK) remains central to the future of the NGDLE despite our best efforts and judgement, and there is a lot of promising thinking around decoupling the pieces, looking at more cohesive integrations through LTIs and APIs, and generally acknowledging there may be life after the LMS, which for many of us who have been waiting for any such sign for 15+ years—that alone is almost enough. The bar is very low in edtech.
I’m pretty tired of LMS bashing; it has pretty much run its course. I still enjoy it from time to time, but I don’t get nearly the thrill I once did back in 2008 or so. Now it’s just kinda depressing. In fact, Leigh Blackall’s recent post on the process of adopting the LMS Canvas at his University captures this pretty well. How long have we been saying this? These discussions make me feel long in the tooth, as do most things in edtech these days. In fact, I was pleasantly surprised by Keegan Long-Wheeler’s presentation at Domains17 wherein he adeptly demonstrated how you can do use LTI integrations from within Canvas. It is premised on two simple tools SSL (via Let’s Encrypt) and Canvas’s redirect tool. The idea being a faculty member can effectively integrate all sorts of small pieces loosely joined cohesively through the LMS.
It builds beautifully on top of Jon Udell’s post from almost a year ago wherein he shares his experience of the LMS while building an app for Hypothes.is, In short, the LTI ecosystem spared him the work of doing a deep dive into the Ruby on Rails framework (what Canvas runs on) to get the Angular framework (what Hypothes.is runs on) talking to one another—the LTI made this possible by simply working like an embeddable script. Yet, for all the promise, there are few examples of anything like a robust integration of various tools from around the web. As cool as these possibilities demonstrated are, they amount to little more than an embeddable script in an LMS post—far from revolutionary in 2017 unless embedding YouTube videos in a WordPress post (achievable more than 10 years ago on the open web) is the end all, be all of future digital learning platforms. I understand much more is possible around the LTI specification, but given there is scarcely a decent tool for integrating WordPress (the most popular publishing engine on the web) into the Canvas ecosystem, I would argue the promise of the LTI is far greater than its adoption and actual use in the two years since the original NGDLE paper was published.
So, what’s the provocation? Beyond the less that overwhelming examples of integrations, my major issue with the vision laid out in the NGDLE article from 2015, was its disregard for how the data supposedly being shared between these systems was being used (and potentially abused) by the various parties involved. Fact is, many of the services focused on personalization and analytics are third-party, commercial services that depend upon data collection for their business model. This opens up a series of very important questions and issues that are effectively glossed over by that article (not to mention the recent follow-up this month that dedicates an entire issue of EDUCAUSE Review to the NGDLE two years on). This stands in stark contrast to a another model for integration of personal data across various systems. The white paper “My Data: A Nordic Model for human-centered personal data management,” authored by Antti Poikola, Kai Kuikkaniemi, and Harri Honko, deals with the same issues facing educational institutions dealing with the NGDLE, but the vision is broader and the focus is not so much on the institution as the individual. It represents the closest thing I have read to a kind of bill of rights around online data:
This model invokes Jon Udell’s discussion of what a personal cyberinfrastructure might look like, and how we would begin to control and manage our personal data in relationship to various entities on the web. In his 2007 talk on the “Disruptive Nature of Technology” he talks about the problem of cohesiveness in the LMS. And while 10 years later he may have a glimmer of hope thanks to technical integrations, what has become of critical importance in the interim is how we manage and control our hosted lifebits (those digital bits we share across various platforms on the web). This question remains crucial, and it is very much the focus of the MyData paper, whereas the NGDLE paper glosses over any discussion of individual privacy, the ethics of collecting student/faculty data, as well as the negotiations around control over one’s personal data. Herein lies the increasingly central issue of edtech, nay our broader digital culture, and is laid out brilliantly by Chris Gilliard’s recent “Pedagogy and the Logic of Platforms,” the most recent contribution to EDUCAUSE”s New Horizons column edited by Mike Caulfield:
The fact that the web functions the way it does is illustrative of the tremendously powerful economic forces that structure it. Technology platforms (e.g., Facebook and Twitter) and education technologies (e.g., the learning management system) exist to capture and monetize data. Using higher education to “save the web” means leveraging the classroom to make visible the effects of surveillance capitalism.
I think the MyData vision of a “human-centered personal data management” approach gets to the heart of what Udell was talking about in 2007 with an eye towards understanding the evolution of these platforms to “capture and monetize” data as Gilliard points out. There is a need for an intervention in the unchecked collection and integration of data across these platforms given data is “the new oil” lubricating the machine of surveillance capitalism.
The NGDLE approach highlights the possibilities of personalization, but plays down the fact it’s premised on the collection of data to create predictive analytics engines by third-party interest. A huge issue for questions surrounding privacy and owning one’s data. This is where one of recent articles about the NGDLE published in the July 3rd issue of EDUCAUSE Review highlights the a particular rhetoric surrounding this still emergent platform. The article in question, namely Stephen Laster’s “Tearing Down Walls to Deliver on the Promise of Edtech:”
While integration might seem to be the concern of IT departments, in truth it has serious implications for teaching and learning. Technologies that live within closed systems create roadblocks for students and instructors as edtech is used to accelerate learner success and faculty efficiency. The free flow of identity, rostering, and learning data, harnessed in service of confident learners and caring faculty, is what allows technology to move us along Bloom’s journey toward mastery learning.
Amongst calls for open standards and a general sense of the limitlessness of transparency, something I generally agree with, it does trouble me when the Chief Digital Officer for McGraw-Hill Publishing is calling for a radical openness when it comes to the “free flow of identity, rostering, and learning data.” One of the travesties of the term open has been it seemingly uncontested goodness when it comes to edtech. I’ve certainly contributed to that problem, but when we are talking about openness of identity and learning data in relationship to students within systems that, despite the hype, have materialized next to nothing—this sounds more like a plan to make sure those various interested parties gain unbridled access to personal data within the NGDLE—the very opposite of the MyData approach being floated in the Nordic Model.
In fact, I think the LMS is dead, not unlike God. The ritual goes on and its re-invented in small, pointless ways to garner a new set of interests and values, it won’t go away, but it also won’t deliver on any of the Silicon Valley-informed Utopian promises outlined in 2015. Rather, in a worst case scenario, the NGDLE offers a way for institutions to more easily extract and share their learning community’s personal data with a wide range of sources, something that should deeply disturb us in the post-Snowden era. But the real kicker is, how do we get anyone to not only acknowledge this process of extraction and monetization (because I think folks have), but to actually feel empowered enough to even care.
The disadvantage of a closed system like BlackBoard is that it’s closed and kind of autistic.
The advantage of a closed system is that it’s closed. Now you can share materials, grades and opinions in safe controlled environment.
Now alas BB is too hackable to be really safe. But the basic idea stays, there’s need for some kind of controlled intemathy.
Our students are not allowed to use patient names nor photos outside the Hospital Patient Information System.
But still you need to discuss cases in a closed environment, even when you’re not using names.
Consider sharing an experience about a male patient with a broken leg that entered ER last night. Even without a name that would be too informative to put it on the open web. Relatives could recognize their family member.
The issue of hyper capitalism getting hold of all our lives and data, is another problem.
Selling our data to the highest bidding dictator is maybe already happening.
Minority Report isn’t far away.
I was at an NGDLE session at EDUCAUSE ELI where people were asked to get into small groups and discuss what was missing. We had a person who had worked on the spec in our group who talked endlessly about all the work that had gone into it. And I don’t doubt it — this stuff is hard.
I asked what they had done towards student control of their data. And the impression I got was that this particular person, who had just detailed the months and months spent working on “all these amazing technologies” that were going to make online stores and exports etc possible had not once thought about student control of their data. Not once.
The flustered reply I got from her was that that was a thing no one was really looking for and besides it was hard.
Well, yeah, it’s hard. What the heck does “Next Generation” stand for if not for trying to do the hard stuff? And as far as people not asking for it, is that our idea of Next Generation? A punch list of what committees asked for?
It’s misnamed. It’s the Adjacent Future LMS. And the adjacent future is basically Blackboard with an Apple Store attached. Which is great, like it honestly might solve some real problems with provisioning niche tech and getting creators paid to create. It’s not nothing. But it felt less than next generation without recentering it on the student.
I am far too late in replying here. The fact you were on the ground for the ELI discussion group for the NGDLE reinforces my reading of the white paper that the idea of student control over their data was not really part of the conceptual framework. And the fact it is hard and would be a total rethinking of the focal point of the LMS does truly make it NG. And I agree it doesn’t mean the NG is necessarily a failure, but just all too familiar with how we are doing business everywhere else on the web.
Can I ask what your biggest critique of my framing of the issues above? Do I overlook some basic issues? I ask because I present a version of this post along with Brian next month, and your writing around new approaches to information literacy (which a truly nextgen LMS might design for) has been really crucial. So, how would you build this thing? 🙂
Hi Jim, thanks for the link. Sorry to be the one banging the old tune, I need to do more to surface the newer criticisms and proposals I make, I don’t want to be ‘that guy’ again.
Over here, in terms of data agency, if not emancipation, we’re having a hard time making proposals. Canvas, for example, has excited managers and administrators with the prospects of data, learning analytics and data dashboards. We’re struggling to make alternative proposals attractive, along the lines of “bring your own account”, “student dashboards”, “a Fitbit for personal learning” and the ability to learn anonymously.
The issue Ron raises, about nurse students inadvertently sharing patient data into the surveillance web, is alarming. I’ve some experience with this were refugee students probably triggered certain surveillance after telling their stories in a forum they *thought* was closed. The only way I could think to mitigate this was by enabling anonymous participation. Impossible in a university environment at present.
The tech paranoia of Black Mirror narratives had injected some maturity into our edtech discourse, all too late sadly. We can all be sorry about that and do our best to use our new wisdoms, humbly.
I really do think the SPLOTs we have been discussing are a very interesting tool for anonymity in certain contexts, and that is a huge piece of what the NGDLE may not even be considering. To what degree do they need to shut off monitoring for folks who do not want it. There is no question the NGDLE framed in the wrong ways becomes part and parcel of a much larger surveillance mechanism already very much entrenched in our culture. And the idea that HE would be a place free from some of that oversight to explore ideas that might be considered dangerous or unsavory would seem far more important to the health of our democracy than pushing towards a series of means and averages that define “good student.”
My take is that nothing in the LMS prevents use of 3rd-party LTI-based tools that empower students to operate autonomously in the LMS environment. For example, the current incarnation of the Hypothesis plugin is just vanilla H: Students create and use their own H accounts.
To me this feels like more than just an embeddable script. The annotation data isn’t in Canvas, only a grade is, and then only if the teacher chooses to apply annotation to an assignment (versus just a reading). The data, which lives in an annotation layer that’s converging on an open standard with multiple interoperable providers, just happens to federate into the course.
Now, since it is inconvenient for teachers to require students to create those accounts, and for students to log into them, everybody wants single sign-on, and that’s sensible. When we deliver SSO, we’ll need to find a way for students to claim and operate H accounts that we initially auto-provisioned by way of the LMS, or to link pre-existing H accounts with Canvas identities. That’s on us, not Canvas, and we’ll do it because we want students to own their lifebits (i.e. annotations on course materials, online relationships formed in those contexts) and have access to them after the course ends.
I like what I see in https://net.educause.edu/ir/library/pdf/eli3035.pdf. From my (outsider) perspective, the LMS can continue to be a key component that supports basic stuff: a syllabus, a roster, an (optional) grade book, and analytics.
I don’t see how that’s necessarily antithetical to course content that’s open, and interaction around that content that accrues to durable identities that exist outside the LMS.
I do see that you have to really, really want to use the LMS that way to keep it in its proper place.
I apologize for the late response to this thoughtful comment. I have no real excuse save that I have been entertaining, traveling, and generally enjoying the Summer life of a wannabe Italian. But I am returning to this post and your comments now because I have to offer a viable critique of some of my ideas here for a conference next month, and I appreciate this feedback immensely. The link there is not resolving, but I totally agree about the need for SSO integration to make the process easier, without it any discussion for scalable tools beyond the LMS is a non-starter. We learned that lesson with Reclaim, and SSO has become a huge part of why schools even considering using us for faculty and student web hosting. What’s more, this piece about making the lifebits more portable is a part of a bigger conversation we started at Domains17 around a vision of archiving one’s lifebits. More on that soon, but I have not forgotten it for sure.
But, and here is where I trying to make sense of things, the call for a learning system to be open mine and possibly sell your data seems to be an issue? I think we can assume that is what the personalized learning vision is all about, and I guess that is where I balk. The NGDLE seems to have no real vision of what a negotiated relationship around that data might be. I am a fan of open, but I think the way it has been hijacked to sell our data to advertisers (however willingly we consign ourselves to perdition) seems to be an element of the design of that system, by placing the vendor at the center rather than the user. Does that make any sense? A default to you owning your x-rays, your medical records, insurance data, electricity usage, but allowing predefined folks to access it for pre-defined reasons and times. I think there is an element to that, but no where does there seem to be a reframing of the terms of that relationship through the “zenlike emptiness” of the NGDLE.
“something that should deeply disturb us in the post-Snowden era.”
I’m fascinated by how thoroughly we’ve accepted the massive compromise of our data.
Belatedly, now, some are focused on (several, rarely all) large firms misusing personal data. But overwhelmingly we not only continue to participate in these systems, we accelerate our use.
Weirdly, scarcely anyone discusses the NSA Five Eyes etc. any longer, at least in the US. Did Democrats just decide it was an acceptable price to pay for the Obama administration’s more favorable acts, and the GOP ditto for Trump? Because we seem to have a bipartisan acceptance of state-run surveillance.
…which might help explain why so many can’t be bothered to get exercised about non-governmental actions in this space.
And this: http://www.socialcooling.com/
Just thought of the idea of how schools using Google Chromebooks give all data to Google. Like Big Brother watching our children as of the first day in school.
Kinda scary, if someone wants to use it in an improper way, it’s all there.